Title: Confirmation of SNPs Associated with Aggressive PCa in a GWA study A genetic predisposition to prostate cancer (PCa) is well established and is the strongest among all common cancers. Inherited sequence variants in a number of genes, each conferring a moderate risk, are believed to collectively underlie the genetic predisposition. To systematically identify these risk variants, we have initiated an ambitious genome-wide association (GWA) study that includes several large and well characterized study populations in Sweden and Johns Hopkins Hospital, totaling >10,000 cases and controls. Thus far, we have completed the 1st stage of this proposed GWA by studying 550K SNPs, including 20K nonsynonymous SNPs, among 500 aggressive cases and 500 controls from a Swedish population (CAPS). To further improve the power of identifying moderate risk SNPs, we propose a study to considerably increase the sample size for a GWA and systematically follow-up a large number of SNPs among independent study populations. We propose four specific aims to test the hypothesis that inherited sequence variants in the genome may increase or modify PCa risk. Aim 1) As the 2nd stage, we will genotype 500K SNPs and a subset of 50K supplement SNPs among an additional 800 aggressive PCa cases and 800 controls from Sweden. A joint association analysis among subjects in stages 1 &2 will be performed to select SNPs for further confirmation. Aim 2) As the 3rd stage, we will test for PCa associations for the ~6,500 SNPs among an additional 2,000 aggressive PCa cases and 1,000 controls from Johns Hopkins Hospital. A combined analysis will be performed for these SNPs among all available subjects to identify SNPs that reach genome-wide significance level. Aim 3) Perform a fine mapping analysis at genomic regions surrounding the genome-wide significant SNPs to identify variants that are most strongly associated with PCa risk among all 3,300 aggressive PCa cases and 2,300 controls. Aim 4) Assess association of the PCa risk variants with the disease progression among 5,000 cases with extensive follow-up information from a Swedish Nationwide Follow-Up study of Localized PCa (FU-study) and 500 matched-pairs of progressors and non-progressors from Johns Hopkins Hospital. The identification of PCa risk variants may impact the understanding, prevention, diagnosis, and treatment of this disease.